The route of drug administration is itself a determinant of the toxicity and addictiveness of a given drug. The smoked route results in far greater morbidity and mortality than all other forms of drug abuse combined with an excess of 400,000 tobacco caused deaths per year, and increasing numbers of deaths caused directly or indirectly by the smoking of heroin, cocaine, marijuana, and methamphetamine (this includes deaths due to HIV infection which is highly prevalent amount drug smokers due to the strong association of this route with sexual activities). Differences in behavioral and physiologic effects of smoked drugs have been the source of considerable speculation as they pertain to developing more effective treatment and prevention strategies. Unfortunately, the pharmacological effects, including mechanisms of reinforcement, have not received thorough study. In part, this is because there have been technical barriers to conducting quantitative research such as well controlled dosing procedures and rapid methods of assessing behavioral and physiologic responses. Research over the past decade at the ARC and in NIDA extramural laboratories has addressed these issues and provided methods that are now being adapted to a range of scientific questions concerning the pharmacology of smoked drugs. Over the past year, we have begun to analyze our extensive data sets from comparative studies of cocaine, heroin, and nicotine, involving both the intravenous and smoked routes of administration (smoked marijuana has also been studied but is reported separately). These studies have examined the arterial drug boli produced by smoking cocaine and nicotine to determine if the smoked route provides physiologically more addicting doses than other routes. These studies showed that certain measures of abuse liability, e.g., liking scores, were similar for heroin, nicotine and cocaine.